WebAssembly example for VAD + Non-streaming ASR (#1284)

This commit is contained in:
Fangjun Kuang
2024-08-24 13:24:52 +08:00
committed by GitHub
parent 1ef8a7a202
commit 537e163dd0
29 changed files with 1281 additions and 70 deletions

View File

@@ -0,0 +1,23 @@
# Introduction
## Download VAD models
Please download
https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/silero_vad.onnx
and put `silero_vad.onnx` into the current directory, i.e., `wasm/vad/assets`.
## Download non-streaming ASR models
Please refer to
https://k2-fsa.github.io/sherpa/onnx/pretrained_models/index.html
to download a non-streaming ASR model, i.e., an offline ASR model.
After downloading, you should rename the model files.
Please refer to
https://k2-fsa.github.io/sherpa/onnx/lazarus/generate-subtitles.html#download-a-speech-recognition-model
for how to rename.
You can find example build scripts at the following address:
https://github.com/k2-fsa/sherpa-onnx/blob/master/.github/workflows/wasm-simd-hf-space-vad-asr.yaml